425

Journal of Andrology, Vol. 10, No. 6, November/December, 1989

Copyright © American Society of Andrology

Ultrastructural Studies on the Development of theBlood-Epididymis Barrier in Immature Rats

ASHOK AGARWAL AND ANITA P. HOFFER

The development of the blood-epididymis barrier inimmature rats (8, 11, 14, 18, and 21 days old) wasexamined with an electron microscope using lanthanumnitrate as an electron dense tracer. A gradual increasein the development of the blood-epididymis barrier wasnoted with age. On Day 8, lanthanum was frequentlydetected in both the intercellular spaces and the lumen.On day 14, no lanthanum penetration into the lumenwas observed in 75% of the junctions in the caput, 40.3%

in corpus, and 30% in cauda epididymidis. On Day 18,only 7.5%, 9%, and 15%, of the junctions in the caput,corpus, and cauda epididymidis, respectively, remainedpermeable to lanthanum. No lanthanum was observedin the lumen of any tubules in the 21-day-old ratepididymis. These findings indicate that the postnataldevelopment of the blood-epididymis barrier is gradual,and that its formation is virtually completed by Day21. As with adult rats, the zonula occludens is theultimate structural component of the blood-epididymisbarrier in immature rats (Agarwal and Hoffer, 1985).

Key words: Blood-epididymis barrier, ultrastructure,immature.

J Androl 1989;10:425-431

Supported by a grant from the Rockefeller Foundation, New

York, NY.

Reprint requests: Dr. Ashok Agarwal, Brigham AndrologyLaboratory, Medical Research Building, Rm. 413, Brigham &

Women’s Hospital, 75 Francis Street, Boston, MA 02115.

Received for publication March 18, 1988; revised April 4, 1989;

accepted for publication May 26, 1989.

From the Harvard Program in Urology andBrigham Andrology Laboratory, Harvard

Medical School and Brigham & Women’sHospital, Boston, Massachusetts

The presence of a blood-tissue barrier restricting

movement of certain substances (e.g. protein, amino

acids, ions) in some body organs has been known

for many years (Waites and Setchell, 1969). The most

important examples are the blood-brain (Reese and

Karnovsky, 1967), blood-thymus (Raviola and

Karnovsky, 1972) and blood-testis (Fawcett et al,

1970; Dym and Fawcett, 1970) barriers. Few

ultrastructural studies on the blood-epididymis

barrier have been published to date (Howards et

al, 1976; Hoffer and Hinton, 1984). In the rat

epididymis, peritubular myoid cell layers, elaboratetight junctions (occluding junction of zonula

occludens), and desmosomes (zonula adhaerens)

serve as a barrier between luminal fluid and blood

capillaries. The epididymal-tight junctions are highly

developed among other epithelial cell contacts

studied. They form a continuous zonule, or gasket,

around the cell, sealing the spaces between the

epithelial cells so that the luminal space and the

intercellular spaces become separate physiological

I

Fig. 1. Electron micrograph of the epididymal epithelium in

the caput of an 8-day-old rat intravascularly perfused with

collidine-buffered glutaraldehyde containing 2% lanthanum

nitrate. Electron-dense tracer is freely distributed along the basal

lamina (not shown) and in the intercellular compartment near

the base of the principal cells.

:..J#{149}., ‘ha

- .

426 Journal of Andrology . November/December 1989 Vol. i

compartments (Friend and Gilula, 1972). Zonula

occludens at the apicolateral surface of the

epididymal epithelium serve as the ultimate

structural component of the rat blood-epididymis

barrier. The flow of intravascularly perfused

lanthanum is not significantly impeded by the

vascular endothelium, peritubular myoid layer, or

other lateral cell surface specializations (Hoffer and

Hinton, 1984).

The present investigation studied for the first

time, the development of the blood-epididymis

barrier in immature rats to determine the exact

morphological location of the structural component

of the blood-epididymis barrier in the head, body,

and tail of the developing epididymis by using an

electron-dense tracer, lanthanum nitrate.

Materials and Methods

Immature Sprague-Dawley rats (Charles RiverLaboratories, Wilmington, MA) 8, 11, 14, 18, and 21 daysold were used. The animals were maintained in accordancewith the guidelines of the Committee on Animals of theHarvard Medical School and those prepared by theCommittee on Care and Use of Animal Resources,National Research Council (DHEW Publication No. (NIH)78-23, revised 1978). Immature rat testes were fixed byantegrade perfusion through the thoracic aorta using 5%

glutaraldehyde with 0.16 M collidine (pH 7.4) containing

2% lanthanum nitrate according to previous methods(Hoffer and Hinton, 1984); except that 2.5 g/100 ml ofpolyvinylpyrollidone (PVP) were dissolved in the collidinebuffer containing lanthanum nitrate by stirring. The pH

of the final fixative (5% glutaraldehyde buffered with 0.16

M collidine containing a final concentration of 2%

lanthanum nitrate and 2.5% PVP) was 6.9. This solution

Fig. 2. Electron micrograph showing

the cauda epididymidis of an 8-day-oldrat. Lanthanum particles can be seen in

the intercellular space near the apical

surface of the principal cells. The absence

of tight junctions at this time intervalallows the lanthanum to penetrate into

the tubular lumen.

1

.

“Jo. 6 BLOOD-EPIDIDYMIS BARRIER IN IMMATURE RATS . Agarwal and Hoffer 427

was routinely filtered through a Buchner funnel(Whatman No. 1) before use. Following a brief rinse incollidine buffer, tissue blocks from the caput, corpus, andcauda were postosmicated and stained en bloc with uranylacetate (Terzakis, 1968) to enhance the contrast ofmembranes and to reveal their trilaminar structure. Afteren bloc staining, the blocks were dehydrated in increasingconcentrations of acetone and embedded in araldite byroutine methods. In some cases, the same lanthanum

concentration was maintained in all subsequent sectionsup to and including the osmium tetroxide; in others,lanthanum was omitted from the immersion solutions to

confirm that the observed path of tracer was causedexclusively by extravasated lanthanum and not by random

diffusion of the tracer from the immersion solutions intothe interstitium.

Sections showing silver to pale-gold interference colors

were cut with a diamond knife on a Sorval MT 6000

microtome. The sections were stained twice with uranylacetate (Watson, 1958) and lead citrate (Venable andCoggeshall, 1965). A Philips 200 or a Jeol lOOs electron

microscope was used to examine the sections. Thepercentage of patent junctions was calculated as follows:

o/p S .o aen -tight junction with lanthanum leaks X 100

Junctions - total number of junctions

A total of 1500 junctions from sections of immature

rat epididymis were counted under a transmission electronmicroscope. In each group, 300 junctions were observed.Chi-square tests were used to compare the percentage

of patency between different age groups (8, 11, 14, 18,and 21 days) and regions of the epididymis (caput, corpus,

and cauda). The null hypothesis was rejected if the chi-square value was � 6.6349 (DF = 1). P values less than

0.05 were considered to be significantly different.

Results

All intercellular contacts between epididymal

epithelial cells, capillary endothelial cells, and

peritubular myoid cells were considered as possible

sites of the blood-epididymis barrier. Each site was

analyzed in the caput, corpus, and cauda epididymidis

to determine whether any variations in the

permeability of the barrier could be detected along

the epididymal duct.

Days 8 and 11

On Day 8, lanthanum was detected in the

intercellular spaces leading to the lumen. The

presence of lanthanum in the basal lamina and in

the intercellular spaces near the apical and the basal

surfaces of the principal cells was observed (Figs.

1 and 2). A paucity of tight junctions, at this time,

allowed the lanthanum to penetrate into the tubular

lumen. Sixty-three percent of the junctions were

patent in the caput, 46% in the corpus, and 32%

Fig. 3. Electron micrograph of the apical cytoplasm of principal

cells in the cauda epididymidis of an 11-day-old rat showing

lanthanum particles in the lumen (arrows).

in the cauda. The most significant decrease in the

percentage of patent junctions was in the caput

region, where it fell from 65% on Day 8 to 3% on

Day 21.

There were no histological differences between

the epididymis of 8 and li-day-old rats. On day

11, 60% of the caput junctions, 36% of the corpus

junctions, and 43% of the cauda junctions were

patent and allowed the passage of lanthanum

particles in the lumen (Fig. 3). Except in the corpus,

there was no signficant difference in the percentage

of patency between 8 and li-day-old rats.

Day 14

In the epididymis of 14-day-old rats, light

microscopic observations of sections from the three

regions revealed an increase in tubular diameter, as

well as an increase in epithelial cell height. Mitotic

division was observed in some of the epithelial cell

428 Journal of Andrology . November/December 1989 Vol. l(

nuclei. The density of cells and fibers in the

intertubular connective tissue had increased, and

tubules were more compactly arranged than in 8-

and li-day-old rats.

In 14-day-old rats, a significant decrease in the

number of patent junctions was observed in the

caput; the corpus and cauda did not reveal any

change. The lanthanum particles were prevented

from reaching the lumen by multiple points of fusion

between the outer leaflets of opposing cell

membranes (Figs. 4 and 5).

Halo cells were first observed at this time. A few

round mitochondria, vesicles, and ribosomes were

present in the cytoplasm. Cytoplasmic processes

extended between the adjacent epithelial cells, but

no junctional complexes were seen between halo

cells and columnar cells.

Day 18 and 21

In the principal cells, the chromatin appeared

diffused as euchromatin in the nucleus. Endoplasmic

reticulum was smooth in the apical cytoplasm, but

granular in the basal region of the cell. Lysosomes

appeared as dense membrane bound sacs. The most

significant change during the third postnatal week

was the appearance of narrow cells in all the three

segments of the epididymis. Basal cells were not

observed. The narrow cells had various features that

set them apart from the rest of the epididymal

epithelium. These cells were clearly identifiable in

light microscopic preparations by their shape and

intense toluidine blue staining. Other components

of the narrow cells included endoplasmic reticulum

consisting of short elements, multi-vesicular bodies,

dense lysosomes, and a small golgi apparatus in the

Figs. 4 and 5. In the caput epididymidis, a signficant decrease in the number of patent junctions is first observed at 14 days

after birth. These electron micrographs of the epididymal epithelium at Day 14 show a tight junction (Fig. 4) and a patent junction

(Fig. 5) in the caput and corpus respectively. Fig. 4: In the junction between the caput epithelial cells, the lanthanum is prevented

from reaching the lumen by multiple points of fusion (arrows) between the outer leaflets of opposing cell membranes. Fig. 5: The

junction between corpus principal cells is still patent and lanthanum can be observed as it extends all the way to the lumen.

a’

No.6 BLOOD-EPIDIDYMIS BARRIER IN IMMATURE RATS . Agarwal and Hoffer 429

Fig. 6. Electron micrograph of the peritubular myoid layer

and basal lamina of the caput epididymidis of the 21-day-oldrat. Extravasated particles of tracer are observed adjacent to

the basal lamina of the epididymal epithelium and appear to

encounter little resistance along the basal and mid portions of

the lateral cell surfaces.

supranuclear cytoplasm. The nucleus was long, and

the chromatin appeared dense compared to that of

surrounding cells.

There was a signficant decrease in the percentage

of patent junctions between 8- and i8-day-old rats,

as well as between 8- and 2i-day-old rats. In the

caput and corpus of 18-day-old rats, about 92% of

the junctions became impermeable to lanthanum

passage (Figs. 6 and 7). In 2i-day-old rats, 97% of

the junctions in the caput and corpus, and 93% in

the cauda became impermeable to lanthanum

passage. In the caput, corpus, and cauda of 2i-day-

old rats, the percentage of patency in tight junctions

did not differ. There was, however, a significant

difference in the percentage of patency between the

caput, corpus, and cauda of i8 and 2i-day-old rats.

The development of the blood-epididymis barrier

was virtually completed by Day 21 (Fig. 8).

Fig. 7. Electron micrograph of the corpus epididymidis of 21-

day-old rat intravascularly perfused with lanthanum. The

passage of tracer particles in the intercellular spaces is obstructedby the zonula occludentes; 97% of the occluding junctions areimpermeable to lanthanum at this time.

Extravasated particles of tracers were observed

adjacent to the basal lamina of the epididymal

epithelium. The particles appeared to encounter little

resistance along the basal and mid portions of the

lateral cell surfaces.

Discussion

The junctional complex is the most elaborate

contact zone between mammalian epithelial cells. It

includes the zonula occludens (tight junction), zonula

adhaerens (intermediate junction), and macula

adhaerens (desmosomes). The junctional complex,

in particular, zonula occludens, is found in epithelia,

which restrict interchange of large molecules

between extracellular and intercellular space (Friend

and Gilula, 1972; Howards et al, 1976; Turner et

al, i980). The zonula occludens of the epididymis

is one of the best developed tight junctions anywhere

8 II 4 IS 21 8 II 14 8 21 8 II 4 lB 21

[ CAPUT I I CORPUS 1 [ CAUDA J

AGE IN DAYS

430 Journal of Andrology . November/December 1989 Vol. 10

FREQUENCY OF EPITHELIAL TIGHT JUNCTIONSIN IMMATURE RAT EPIDIDYMIS

Fig. 8. Frequency of epithelial tight junctions in immaturerat epididymis, ‘significantly different from 8-day interval at

P < 0.01, 1’not significantly different from 8-day interval at P

<0.01, ‘Patent junctions =

Occluding junctions with lanthanum leaks X 100

total number of junctions

in mammalian tissues (Friend and Gilula, 1972). The

ultimate site of the blood-epididymis barrier differs

from the blood-tissue barriers of other organs, such

as the thymus (Raviola and Karnovsky, 1972), the

brain (Reese and Karnovsky, 1967), and the testis

(Fawcett et al, 1970; Dym and Fawcett, i970). In

the epididymis, the zonula occludens near the

luminal surface of the epididymal epithelium are

exclusively responsible for the maintenance of the

blood-epididymis barrier, since neither the capillary

endothelium nor the peritubular myoid layer

significantly impedes the flow of tracers towards

the epididymal lumen (Hoffer and Hinton, i984).

Although ultrastructural studies of the blood-

epididymis barrier in adult rats have been reported

earlier from this laboratory (Hoffer and Hinton,

1984), the development 0f the blood-epididymis

barrier in the mammalian epididymis is described

here for the first time.

The results of the present study indicate that the

development of the blood-epididymis barrier is

virtually complete by the end of the third post-natal

week. Even though factors controlling the overall

completion of the blood-epididymis barrier by this

time are not completely understood, it is noteworthy

that the blood-epididymis barrier is completed at

approximately the same time as the blood-testis

barrier. Vitale et a! (1973) have demonstrated that

in the rat testes, the blood-testis barrier is established

between post-natal Days 16 and i9 in close temporal

correlation with the appearance of junctional

complexes between Sertoli cells, the onset of fluid

secretion by the seminiferous epithelium, the

stratification of germinal epithelium, and the

development of a lumen in the seminiferous tubule.

Using acridine dyes, Kormano (i967) found that the

permeability barrier in testes develops gradually

during the first 20 post-natal days. During the third

post-natal week there is a steep rise in serum

gonadotrophins and testosterone (Miyachi et al,

1973; Swerdloff et al, 1971), epididymal secretion

of glycerylphosphorylcholine, sialic acid and alkaline

phosphatase (Goyal and Dhingra, 1974; Setty and

Jehan, i977), and ABP in the principal cells of the

caput epididymidis (Hansson et al, i974; White et

al, 1982). These changes reflect the early activation

of pituitary and Leydig cells, which may play a role

in initiating sexual maturation and increased

impermeability of most junctions throughout the

epididymis by Day 2i. Of interest is the observation

that many of the intercellular junctions in the corpus

and cauda are already tight by post-natal Day 8.

In the rat brain, the blood-brain barrier to protein

develops early in fetal life and is well-established

by birth (Olsson et al, 1968). The exact time when

most (-2/3) intercellular junctions in the cauda

epididymidis first became impermeable to

lanthanum is not known, since the development of

the blood-epididymis barrier in animals less than 8-

days-old has not been examined to date.

Another intriguing observation is that differences

in barrier development rate between epididymal

zones can be noted as early as the first 2 post-natal

weeks. In fact, by post-natal Day 8, the proportion

of intercellular junctions that are permeable to

lanthanum in the caput, corpus, and cauda varies

from approximately 2/3 to i/2 to 1/3, respectively.

By week 2, there is a disproportionately large drop

in the number of patent junctions in the caput,

compared to the more distal regions of the

epididymal duct. Declines of comparable magnitude

do not take place until Day 18 in the corpus and

cauda. Early maturation of the blood-epididymis

barrier in the caput is consistent with the

observation of Sun and Flickinger (i979) that

histological maturation occurs earlier in the proximal

regions of the epididymis than in the distal ones.

The reasons for the initial differences in maturity

and the subsequent differential rate of blood-

epididymis barrier development, however, are not

clear. In the adult rat, maintenance of the initial

segment requires more intraluminal testosterone

No.6 BLOOD-EPIDIDYMIS BARRIER IN IMMATURE RATS . Agarwal and Hoffer 431

than the cauda (Fawcett and Hoffer, 1979), but

whether this is relevant to the neonatal epididymis

is not known. The greater immaturity of the caput

blood-epididymis barrier on Day 8, relative to the

distal gential duct, is consistent with this possibility.

It is also noteworthy that during post-natal days

0-14, there is a steady decline of all endogenous

steroids except testosterone and androstanediol,

whereas, during days 10-15, the testicular level of

androstanediol increases significantly until it

emerges as the predominant endogenous androgen

of testes in rats 15 days and older (Tappanien et

al, 1984). It is tempting to speculate that blood-

epididymis barrier development in the prepubertal

epididymis, especially the caput, has a specific need

for androstanediol or some other unknown factor(s),

and that it cannot mature until this factor(s) is

available around the second post-natal week. An

alternate hypothesis is that, since the lumen of the

seminiferous tubules is formed more or less

concurrently with the establishment of the blood-

epididymis barrier, the intraluminal presence of

testicular secretory products in the epididymis may

be required for completion of blood-epididymis

barrier development.

References

Agarwal A, Hoffer AP. Ultrastructural studies on the

development of the blood-epididymis barrier in immature

rats. J Androl Suppi 1985;6:90.

Dym M. Fawcett DW. The blood-testis barrier in the rat and

the physiological compartmentation of the seminiferous

epithelium. Biol Reprod 1970;3:308-326.

Fawcett DW, Hoffer, AP. Failure of exogenous androgen toprevent regression of the initial segments of the rat

epididymis after efferent duct ligation or orchiectomy. BiolReprod 1979;20:162-181.

Fawcett DW, Leak LC, Heidger PM. Electron microscopic

observations on the structural components of the blood-testis

barrier. J Reprod Fertil Suppl 1970;10:105-122.

Friend DS, Gilula NB. Variations in tight and gap junctions in

mammalian tissues. J Cell Biol 1972;53:758-776.

Goyal HO, Dhingra LD. Histochemical study on the postnataldistribution of alkaline phosphatase in the epididymis of

buffalo (Bubalis bubalis): from birth to one and half years.

Acta Anat 1974;88:574-579.

Hansson V. Trygstad 0, French FS, McLean WS, Smith AA,

Tindall DJ, Weddington SC, Petrusz F, Nayfeh SM, RitzenEM. Androgen transport and receptor mechanisms in testis

and epididymis. Nature (London) 1974;250:387-391.

Hoffer AP, Hinton BT. Morphological evidence for a blood-

epididymis barrier and the effects of gossypol on its integrity.

Biol Reprod 1984;30:991-1004.Howards SS, Jessee SJ, Johnson AL. Micropuncture studies of

the blood-seminiferous tubule barrier. Biol Reprod

1976;14:264-269.

Kormano M. Dye permeability and alkaline phosphatase activity

of testicular capillaries in the postnatal rat. Histochemie1967;9:327-338.

Miyachi Y, Nieschlag E, Lipsett MV. The secretion of

gonadotrophins and testosterone by the neonatal male rats.

Endocrinology 1973;92:1-5.

Olsson Y, Klatzo I, Sourander P. Steinwall 0. Blood-brain barrier

to albumin in embryonic newborn and adult rats. Acta

Neuropathol 1968;10:117-122.

Raviola E, Karnovsky MJ. Evidence for a blood-thymus barrier

using electron-opaque tracers. J Exp Med 1972;136:466-498.

Reese TS, Karnovsky MJ. Fine structural localization of a blood-

brain barrier to exogenous peroxidase. J Cell Biol

1967;34:207-217.

Setty BS, Jehan Q. Functional maturation of the epididymis inthe rat. J Reprod Fertil 1977;49:317-322.

Sun EL, Flickinger CJ. Development of cell types and of regional

differences in the postnatal rat epididymis. Am J Anat

1982;203:273-284.

Swerdloff RS, Walsh PC, Jacobs HS, Odell WD. Serum LH andFSH during sexual maturation in the male rat: Effect ofcastration and cryptorchidism. Endocrinology 1971;88:120-

128.Tapanainen J, Kuopio T, Pelliniemi U, Huh taniomi I. Rat testicular

endogenous steroids and number of Leydig cells betweenthe fetal period and sexual maturity. Biol Reprod 1984;-

31:2017-1035.Terzakis JA. Uranyl acetate, a stain and a fixative. J Ultrastruct

Res 1968;22:168-184.

Turner IT, D’Addario DA, Howards SS.lHj3-0-methyl-D-

glucose transport from blood into the lumina of the

seminiferous and epididymal tubules in intact and

vasectomized hamsters. J Reprod Fertil 1980; 60:285-289.

Venable JH, Coggeshall R. A simplified lead citrate stain for use

in electron microscopy. I Cell Biol 1965;25:407.

Vitale R, Fawcett DW, Dym M. The normal development ofthe blood-testis barrier and the effects of clomiphene and

estrogen treatment. Anat Rec 1973;176:333-344.

Waites CMH, Setchell BP. Physiology of the testis, epididymis

and scrotum. In: McLaren A, Ed. Advances in Reproductive

Physiology, London: Logos Press, 1969.

Watson ML. Staining of tissue sections for electron microscopy

with heavy metals. J Biophys Biochem Cytol 1958;4:475.

White IC, Huang VS. Tres LL, Kierszenbaum AL. Structural

and functional aspects of cultured epididymal epithelial cellsisolated from pubertal rats. J Reprod Fertil 1982;66:475-484.

Notice to Copiers

Authorization to photocopy items for internal or personal use, or the

internal or personal use of specific clients, is granted by J. B. Lippincott

Company for libraries and other users registered with the Copyright

Clearance Center (CCC) Transactional Reporting Service, provided that

the base fee of $00.80 per copy, plus $00.05 per page is paid directly

to CCC, 21 Congress St., Salem, MA 01970. 0i96-3635/89 $00.80 +

$00.05.